[Mesa-dev] [PATCH v2 02/52] intel/fs: Be more explicit about our placement of [un]zip

[Jason Ekstrand]

Before, we were careful to place the zip after the last of the split
instructions but did unzip on-demand.  This changes things so that the
unzips go before all of the split instructions and the unzip comes
explicitly after all the split instructions.  As a side-effect of this
change, we now emit the split instruction from highest SIMD group to
lowest instead of low to high.  We could have kept the old behavior, but
it shouldn't matter and this made the code easier.

Cc: mesa-stable at lists.freedesktop.org
---
 src/intel/compiler/brw_fs.cpp | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

diff --git a/src/intel/compiler/brw_fs.cpp b/src/intel/compiler/brw_fs.cpp
index dc29765..0eebc70 100644
--- a/src/intel/compiler/brw_fs.cpp
+++ b/src/intel/compiler/brw_fs.cpp
@@ -5190,6 +5190,7 @@ fs_visitor::lower_simd_width()
 
          assert(!inst->writes_accumulator && !inst->mlen);
 
+         exec_node * const after_inst = inst->next;
          for (unsigned i = 0; i < n; i++) {
             /* Emit a copy of the original instruction with the lowered width.
              * If the EOT flag was set throw it away except for the last
@@ -5197,7 +5198,7 @@ fs_visitor::lower_simd_width()
              */
             fs_inst split_inst = *inst;
             split_inst.exec_size = lower_width;
-            split_inst.eot = inst->eot && i == n - 1;
+            split_inst.eot = inst->eot && i == 0;
 
             /* Select the correct channel enables for the i-th group, then
              * transform the sources and destination and emit the lowered
@@ -5209,11 +5210,11 @@ fs_visitor::lower_simd_width()
                split_inst.src[j] = emit_unzip(lbld.at(block, inst), inst, j);
 
             split_inst.dst = emit_zip(lbld.at(block, inst),
-                                      lbld.at(block, inst->next), inst);
+                                      lbld.at(block, after_inst), inst);
             split_inst.size_written =
                split_inst.dst.component_size(lower_width) * dst_size;
 
-            lbld.emit(split_inst);
+            lbld.at(block, inst->next).emit(split_inst);
          }
 
          inst->remove(block);
-- 
2.5.0.400.gff86faf